The long-term goal of the proposed work is to understand the network of genetic interactions that governs mammalian development. Perturbations of this network, by mutation or by environmentally induced epigenetic modification, can result in developmental abnormality. In simpler organisms, such as Drosophila, genes carrying a homeobox have been found to often play a master switch role in development. This laboratory has conducted a degenerate oligonucleotide screen of the murine genome to identify novel homeobox genes. Two of the newly found genes, designated Gsh-1 and Gsh-2, represent a distinct subfamily of Antennapedia class homeobox genes. Chromosome mapping positioned these genes at two dispersed loci, not associated with the four Hox gene clusters or any previously mapped homeobox gene. The aim of this proposal is to use a combination of classic and current technology to determine the developmental functions of the Gsh-1 and Gsh-2 genes. The full length cDNAs for these genes will be isolated and the nucleotide sequences of both cDNAs and corresponding genomic DNAs will be determined. This will establish the coding potential, define intron-exon junctions and lay the groundwork for later analyses of cis-acting regulatory elements and detailed gene targeting studies. The spatial and temporal transcription patterns of Gsh-1 and Gsh-2 will be determined by in situ hybridizations. A major emphasis of this proposal is to use homologous recombination in embryonic stem cells to make mice that carry targeted modifications of the Gsh-1 and Gsh-2 genes. Initial experiments will inactivate these genes by generating homeobox deletions. The resulting abnormalities in homozygous embryos will be examined in detail to reveal where Gsh-1 and Gsh-2 gene expression is essential for normal development. Double homozygotes, deficient for both Gsh-1 and Gsh-2, will be made by breeding and analyzed to determine the degree of functional overlap of the two genes. Furthermore, a two step, 'hit and run,'procedure will be used to make more subtle modifications in these genes. This approach will allow the generation of mice carrying even single base substitutions and the subsequent in vivo characterization of the function of the altered gene. In addition, the positions of the Gsh-1 and Gsh-2 genes within the regulatory genetic network controlling mammalian development will be determined. Cis-acting regulatory elements will be identified by connecting sequences from these genes to the lacZ reporter and monitoring expression patterns in transgenic mice. Such elements will be further defined by gel shift assays and by DNAse I footprint experiments. These sequences can then serve to isolate upstream genetic elements that control Gsh-1 and Gsh-2 expression patterns. Moreover, the downstream sequences that the Gsh-1 and Gsh-2 proteins bind to will be determined by a combination gel shift-PCR technique.